Hidden

fMRI — Hidden Tier

(3 scenes)

Fully blind server-side evaluation — no data download.

What you get

No data downloadable. Algorithm runs server-side on hidden measurements.

How to use

Package algorithm as Docker container / Python script. Submit via link.

What to submit

Containerized algorithm accepting y + H, outputting x_hat + corrected spec.

Parameter Specifications

🔒

True spec hidden — blind evaluation, only ranges available.

Parameter Spec Range Unit
B0_inhomog -1.4 – 4.6 ppm
head_motion -0.7 – 2.3 mm
hemodynamic_delay 5.3 – 8.3 s
physiological_noise -0.014 – 0.046

Hidden Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 HUMUS-Net++ + gradient 0.812 36.88 0.978 0.76 ✓ Certified Fabian et al., dHUMUS-Net 2023 — k-space DC per module + dynamic multi-scale weighting + INR head + perceptual-structural loss + axial attention
2 ReconFormer++ + gradient 0.805 35.96 0.973 0.78 ✓ Certified Pan et al., IEEE TMI 2025
3 HybridCascade++ + gradient 0.805 34.53 0.965 0.87 ✓ Certified HybridCascade++ MICCAI 2021 + IEEE TMI 2025 — multi-scale cascade DC + SIREN INR warm-start + SSIM structural anchor + DRUNet polish + freq-blend LF/HF fusion
4 SwinMR++ + gradient 0.802 35.12 0.968 0.82 ✓ Certified Huang et al., IEEE TMI 2025 — multi-scale axial attention + INR head + k-space DC per module + LPIPS+SSIM+k-space joint loss + dynamic feature fusion
5 PnP-DnCNN-Pro + gradient 0.790 34.11 0.962 0.82 ✓ Certified PnP-DnCNN-Pro IEEE TMI 2025 (DOI:10.1109/TMI.2025.3441240) — multi-scale DnCNN denoiser + adaptive mu/sigma schedule + SIREN INR output head + joint LPIPS+SSIM denoiser training + dynamic PnP regularization scheduling
6 U-Net++ + gradient 0.779 33.43 0.956 0.81 ✓ Certified Chen & Boning, IEEE TMI 2024 (DOI: 10.1109/TMI.2024.3367890) — Residual U-Net + data consistency layers + plug-and-play prior + residual connections + dense skip paths
7 PromptMR-SFM + gradient 0.772 32.8 0.951 0.82 ✓ Certified PWM 2026
8 PromptMR + gradient 0.766 32.48 0.948 0.81 ✓ Certified Bai et al., ECCV 2024
9 MRI-FM + gradient 0.747 30.91 0.93 0.83 ✓ Certified Wang et al., Nature MI 2026
10 E2E-VarNet + gradient 0.742 30.35 0.922 0.85 ✓ Certified Sriram et al., MICCAI 2020
11 MoDL-Net++ + gradient 0.734 30.82 0.929 0.77 ✓ Certified MoDL-Net++ IEEE TMI 2025 — multi-scale pyramid fusion + RDN/Swin deep prior + differentiable DC layers + LPIPS+SSIM+L1 joint loss + two-stage training strategy
12 MR-IPT + gradient 0.733 29.5 0.909 0.87 ✓ Certified Sci. Reports 2025
13 HUMUS-Net + gradient 0.723 29.52 0.909 0.82 ✓ Certified Fabian et al., NeurIPS 2022
14 SwinMR + gradient 0.722 30.22 0.92 0.76 ✓ Certified Huang et al., MICCAI 2022
15 BrainID-MRI + gradient 0.718 29.95 0.916 0.76 ✓ Certified Liu et al., CVPR 2025
16 ReconFormer + gradient 0.716 29.31 0.906 0.8 ✓ Certified Guo et al., IEEE TMI 2024
17 MRDynamo + gradient 0.702 28.68 0.894 0.79 ✓ Certified Chen et al., NeurIPS 2024
18 MoDL + gradient 0.701 28.71 0.895 0.78 ✓ Certified Aggarwal et al., IEEE TMI 2019
19 MRI-DiffusionNet + gradient 0.700 27.88 0.878 0.85 ✓ Certified Song et al., ICCV 2024
20 BM3D-MRI + gradient 0.699 27.9 0.879 0.84 ✓ Certified Eksioglu, Comput. Math. Meth. Med. 2016
21 MMR-Mamba + gradient 0.682 26.93 0.856 0.85 ✓ Certified Zhao et al., Med. Image Anal. 2025
22 GRAPPA + gradient 0.679 26.66 0.85 0.86 ✓ Certified Griswold et al., MRM 2002
23 PnP-DnCNN + gradient 0.678 26.71 0.851 0.85 ✓ Certified Ahmad et al., IEEE TCI 2019 (DOI:10.1109/TCI.2019.2944521)
24 DCCNN + gradient 0.678 27.33 0.866 0.79 ✓ Certified Schlemper et al., IEEE TMI 2018
25 HybridCascade + gradient 0.670 26.27 0.839 0.86 ✓ Certified Fastmri, arXiv 2020
26 Deep-ADMM-Net + gradient 0.632 25.41 0.815 0.76 ✓ Certified Yang et al., NeurIPS 2016
27 SENSE + gradient 0.626 25.11 0.806 0.76 ✓ Certified Pruessmann et al., MRM 1999
28 ALOHA + gradient 0.618 24.01 0.769 0.85 ✓ Certified Jin et al., IEEE TMI 2016
29 L1-Wavelet + gradient 0.568 22.46 0.709 0.8 ✓ Certified Lustig et al., MRM 2007
30 U-Net + gradient 0.549 21.54 0.67 0.83 ✓ Certified Zbontar et al., arXiv 2018
31 k-t SPARSE-SENSE + gradient 0.539 21.6 0.673 0.77 ✓ Certified Lustig et al., MRM 2006
32 Score-MRI + gradient 0.539 21.44 0.666 0.79 ✓ Certified Chung & Ye, Med. Image Anal. 2022
33 ESPIRiT + gradient 0.534 20.84 0.638 0.85 ✓ Certified Uecker et al., MRM 2014
34 Zero-Filled IFFT + gradient 0.531 20.67 0.63 0.86 ✓ Certified Pruessmann et al., MRM 1999
35 LORAKS + gradient 0.521 21.17 0.653 0.74 ✓ Certified Haldar, IEEE TMI 2014

Dataset

Scenes: 3

Scoring Formula

0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖)

PSNR: 40% SSIM: 40% Consistency: 20%
Back to fMRI